Multisensor estimation of vegetation characteristics

The case for a multisensor approach to estimate and monitor vegetation characteristics has been well-established. SAR sensors have shown promise in not only classifying vegetation types but also in estimating parameters such as biomass, canopy height, and diameter at breast height (dbh). The accuracy with which vegetation types can be classified and the above parameters estimated can be significantly improved by using data from other optical sensor systems such as color-infrared (IR) imagery and satellite photography. We have obtained contemporaneous and coregistered SIR-C SAR and airborne color-IR images as well as satellite photographs of a forested area in New Hampshire. Bayesian classification technique is being investigated in order to classify vegetation into broad classes. Inversion algorithms are also being developed for estimating specific vegetation parameters once broad classes have been delineated. The added benefit of integrating optical sensor data with the SAR imagery is being studied in terms of classification and estimation accuracy.